In the art of growing semiconductor materials and particularly infrared modulator and window materials, it is manifestly desirable to grow either in bulk polycrystalline form or by epitaxial deposition certain selected binary or ternary compound semiconductors with near-perfect stoichiometry and with a minimum of defects (and thus vacancies) in the crystal lattice of the given material. As is well known, the stoichiometry, impurities and native defects in the grown crystal determine the electrical and optical properties of the material. And, in some instances, the optimum electrical and optical properties of the grown material correspond directly to an optimum in stoichiometry and a minimum in impurities and lattice defects in the grown crystal. For example, a minimum optical absorption coefficient for certain II-VI semiconductor compounds corresponds directly to an optimum in the stoichiometry and a minimum in impurities and lattice defects in the grown crystal. For a further discussion of the relationship of stoichiometry, impurities and native defects of certain II-VI compound semiconductors, e.g. CdTe, to their optical properties, reference may be made to an article by A. L. Gentile et al entitled "A Thermal Annealing Procedure For the Reduction of 10.6 .mu.m Optical Losses in CdTe", Material Research Bulletin, Vol. 8, pp 523-532, 1973, Pergamon Press, Inc.
Many different types of bulk polycrystalline and epitaxial single crystal growth processes have been used in the past to grow selected II-VI compound semiconductors, and particularly the following compound semiconductors with which the present invention is primarily concerned:
Zinc Selenide (ZnSe) PA1 Zinc Sulphide (ZnS) PA1 Zinc Telluride (ZnTe) PA1 Cadmium Sulphide (CdS) PA1 Cadmium Selenide (CdSe) PA1 Cadmium Telluride (CdTe)
Among these processes are included certain vapor transport processes which are carried out at approximately atmospheric pressure and using known inert carrier gases to combine selected reactants at a chosen location in the formation of the desired compound semiconductor. The location could, for example, be one surface of a graphite crucible where bulk polycrystalline compound semiconductors are formed, or it could be a selected monocrystalline substrate upon which the material is epitaxially deposited.
While the above vapor transport processes have generally proven satisfactory in the formation of compound semiconductive materials used for certain types of optical applications, they are inherently limited, purity wise, by the open-tube nature of the system and are therefore inadequate for producing suitable high purity compound semiconductors for other special types of optical applications.
One process which has been used to improve upon the stoichiometry and purity levels of crystals grown by the above described vapor transport process employs a sealed tube which is located in a furnace having a predetermined temperature profile. Such profile is closely controlled with respect to the source and grown materials which are located, respectively, at opposite ends of the sealed tube. In this process, the differential pressures produced by the controlled thermal gradient in the tube are responsible for the transport of elemental vapors of the source material to the other end of the tube where they recombine and are deposited either on a crucible or a selected substrate.
While this closed tube process improves upon the stoichiometry and purity levels of the grown crystals relative to those produced by the above open-tube system, it is nevertheless a relatively slow process and is still limited, purity-wise, as a result of the completely closed nature of the system. These two disadvantages have been either partially or completely eliminated by the present invention to be described below.